Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Propagation of chaos for the homogeneous Boltzmann equation with moderately soft potentials

Nicolas Fournier, Stéphane Mischler

TL;DR

This work proves propagation of chaos for the homogeneous Boltzmann equation with moderately soft potentials $\gamma\in(-2,0)$ by showing that the Kac $N$-particle system converges to the Boltzmann solution $f_t$ as $N\to\infty$, with the limit carrying entropic chaos properties. The key technical mechanism is the decay of the $N$-particle Fisher information, inherited from a recent Boltzmann analysis of Imbert, Silvestre and Villani, which yields uniform control of singular interactions and enables compactness and limit passage. The authors build a rigorous path from the Kac system to a nonlinear stochastic process $V_t$ solving a nonlinear martingale problem $MP(f_0)$, establish uniqueness for time-integrable Fisher information, and obtain chaos in both trajectory and entropic senses. This provides a robust derivation of Boltzmann dynamics from particle systems in the physically relevant soft-potential regime, and clarifies the role of Fisher information as a tool to manage singular interactions in kinetic theory.

Abstract

We show that the Kac particle system converges, as the number of particles tends to infinity, to the solution of the homogeneous Boltzmann equation, in the regime of moderately soft potentials, $γ\in (-2,0)$ with the common notation. This proves the propagation of chaos. We adapt the recent work of Imbert, Silvestre and Villani, to show that the Fisher information is nonincreasing in time along solutions to the Kac master equation. This estimate allows us to control the singularity of the interaction.

Propagation of chaos for the homogeneous Boltzmann equation with moderately soft potentials

TL;DR

This work proves propagation of chaos for the homogeneous Boltzmann equation with moderately soft potentials by showing that the Kac -particle system converges to the Boltzmann solution as , with the limit carrying entropic chaos properties. The key technical mechanism is the decay of the -particle Fisher information, inherited from a recent Boltzmann analysis of Imbert, Silvestre and Villani, which yields uniform control of singular interactions and enables compactness and limit passage. The authors build a rigorous path from the Kac system to a nonlinear stochastic process solving a nonlinear martingale problem , establish uniqueness for time-integrable Fisher information, and obtain chaos in both trajectory and entropic senses. This provides a robust derivation of Boltzmann dynamics from particle systems in the physically relevant soft-potential regime, and clarifies the role of Fisher information as a tool to manage singular interactions in kinetic theory.

Abstract

We show that the Kac particle system converges, as the number of particles tends to infinity, to the solution of the homogeneous Boltzmann equation, in the regime of moderately soft potentials, with the common notation. This proves the propagation of chaos. We adapt the recent work of Imbert, Silvestre and Villani, to show that the Fisher information is nonincreasing in time along solutions to the Kac master equation. This estimate allows us to control the singularity of the interaction.
Paper Structure (13 sections, 12 theorems, 135 equations)

This paper contains 13 sections, 12 theorems, 135 equations.

Table of Contents

  1. Introduction and main results
  2. The homogeneous Boltzmann equation
  3. The Kac particle systems
  4. Propagation of chaos
  5. Main result
  6. Strategy of the proof
  7. Plan of the paper
  8. The Fisher information
  9. Reminder about the Boltzmann equation
  10. Kac $N$-particle system
  11. The nonlinear stochastic equation
  12. From the Kac system to the nonlinear stochastic equation
  13. Entropic propagation of chaos

Key Result

Theorem 1.1

Consider a kernel $B$ satisfying hyp0-hyp1 with $\gamma \in (-2,0)$ and a nonnegative initial condition $f_0$ satisfying the normalization and with finite Fisher information. Denote by $f \in C([0,\infty);L^1({\mathbb{R}}^3))$ the unique weak solution to the Boltzmann equation be with nonincreasing Fisher information and which which satisfies eq:normalizationf. For any $N\geq 2$, there exists a $

Theorems & Definitions (26)

  • Theorem 1.1
  • Remark 1.2
  • Lemma 2.1
  • proof
  • Proposition 2.2
  • proof
  • Lemma 2.3
  • proof
  • Definition 3.1
  • Theorem 3.2
  • ...and 16 more